Thanks to Elena and Barbara's patient help, it appears lowering (by
editting config/intel-flags) the optimisation to -O1 is required for icc
11.0 (Build 20081105 Package ID: l_cproc_p_11.0.074) running under
Scientific Linux:
$ cat /etc/redhat-release ; uname -a
Scientific Linux SL release 5.2 (Boron)
Linux redqueen.rcs.manchester.ac.uk 2.6.18-238.1.1.el5 #1 SMP Tue Jan 18
18:49:37 EST 2011 x86_64 x86_64 x86_64 GNU/Linux
Many thanks, Michael
On Fri, 2011-06-24 at 22:43 +0400, Elena Pourmal wrote:
> Michael,
>
> In 1.8.7 we enabled -O3 optimization for C Intel compiler. May be the higher
> level of optimization doesn't work on your machine (errors sound familiar,
> but I may be wrong)
>
> Could you please set CC to "icc -O2" and recompile the library, or fix the
> config/intel-flags file (search 1.8.7 string to find the right place).
>
> Thank you!
>
> Elena
> On Jun 24, 2011, at 4:01 PM, michael bane wrote:
>
> > My hdf5 1.8.7 appears to build okay but my make check fails. This is on a
> > Sci Linux box:
> > uname -a
> > Linux redqueen.rcs.manchester.ac.uk 2.6.18-238.1.1.el5 #1 SMP Tue Jan 18
> > 18:49:37 EST 2011 x86_64 x86_64 x86_64 GNU/Linux
> >
> > I attach the std out+err from my configure and make steps. And here is the
> > detailed error
> >
> > [mccssmb2@redqueen hdf5-1.8.7]$
> > /software/hdf/5/1.8.7/hdf5-1.8.7/test/.libs/lt-testhdf5 -vv -o select 2>&1
> > | less
> >
> > For help use: /software/hdf/5/1.8.7/hdf5-1.8.7/test/.libs/lt-testhdf5 -help
> > Linked with hdf5 version 1.8 release 7
> > Testing -- Configure definitions (config)
> > Testing -- Encoding/decoding metadata (metadata)
> > Testing -- Checksum algorithm (checksum)
> > Testing -- Ternary Search Trees (tst)
> > Testing -- Memory Heaps (heap)
> > Testing -- Skip Lists (skiplist)
> > Testing -- Reference Counted Strings (refstr)
> > Testing -- Low-Level File I/O (file)
> > Testing -- Generic Object Functions (objects)
> > Testing -- Dataspaces (h5s)
> > *** UNEXPECTED RETURN from H5Fopen is -1 line 188 in th5s.c
> > HDF5-DIAG: Error detected in HDF5 (1.8.7) thread 0:
> > #000: H5F.c line 1522 in H5Fopen(): unable to open file
> > major: File accessability
> > minor: Unable to open file
> > #001: H5F.c line 1211 in H5F_open(): unable to open file: time = Fri Jun 24
> > 12:41:23 2011
> > , name = 'th5s.h5', tent_flags = 0
> > major: File accessability
> > minor: Unable to open file
> > #002: H5FD.c line 1086 in H5FD_open(): open failed
> > major: Virtual File Layer
> > minor: Unable to initialize object
> > #003: H5FDsec2.c line 348 in H5FD_sec2_open(): unable to open file: name =
> > 'th5s.h5', errno = 2, error mess
> > age = 'No such file or directory', flags = 0, o_flags = 0
> > major: File accessability
> > minor: Unable to open file
> > ***cannot open the pre-created H5S_MAX_RANK test file (th5s.h5)
> > Testing -- Dataspace coordinates (coords)
> > Testing -- Shared Object Header Messages (sohm)
> > Testing -- Attributes (attr)
> > Testing -- Selections (select)
> > large cube read from small cube has bad data! Line=2015
> > large cube read from small cube has bad data! Line=2015
> > large cube read from small cube has bad data! Line=2015
> > large cube read from small cube has bad data! Line=2015
> > large cube read from small cube has bad data! Line=2015
> > large cube written from small cube has bad data! Line=2256
> > large cube written from small cube has bad data! Line=2256
> > large cube written from small cube has bad data! Line=2256
> > large cube written from small cube has bad data! Line=2256
> > large cube written from small cube has bad data! Line=2256
> > large cube read from small cube has bad data! Line=2015
> > large cube read from small cube has bad data! Line=2015
> > {& much more of same}
> >
> >
> > Many thanks, Michael
> > <mkb_make.txt><mkb_configure.txt>_______________________________________________
> > Hdf-forum is for HDF software users discussion.
> > [email protected]
> > http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
>
>
> _______________________________________________
> Hdf-forum is for HDF software users discussion.
> [email protected]
> http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
--
Dr. Michael K. Bane
Snr Research Apps & Collabs Consultant
IT Services for Research
Univ. of Manchester M13 9PL
http://www.rcs.manchester.ac.uk
http://www.twitter.com/mkbane_mcr
_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
